Interviewing in Social Science Research

A Relational Approach

Lee Ann Fujii

• Publisher: Routledge
• Year: 2017
• eBook ISBN: 9781135015381
• Topic: Politics & International Relations
• Subtopic: Political History & Theory
• Index: Politics & International Relations

What is interviewing and when is this method useful? What does it mean to select rather than sample interviewees? Once the researcher has found people to interview, how does she build a working relationship with her interviewees? What should the dynamics of talking and listening in interviews be? How do researchers begin to analyze the narrative data generated through interviews?

Lee Ann Fujii explores the answers to these inquiries in Interviewing in Social Science Research, the latest entry in the Routledge Series on Interpretive Methods. This short, highly readable book explores an interpretive approach to interviewing for purposes of social science research. Using an interpretive methodology, the book examines interviewing as a relational enterprise. As a relational undertaking, interviewing is more akin to a two-way dialogue than a one-way interrogation. Fujii examines the methodological foundations for a relational approach to interviewing, while at the same time covering many of the practical nuts and bolts of relational interviewing. Examples come from the author's experiences conducting interviews in Bosnia, Rwanda, and the United States, and from relevant literatures across a variety of social scientific disciplines. Appendices to the book contain specific tips and suggestions for relational interviewing in addition to interview excerpts that give readers a sense of how relational interviews unfold.

This book will be of great value to graduate students and researchers from across the social sciences who are considering or planning to use interviews in their research, and can be easily used by academics for teaching courses or workshops in social science methods.

1
WHAT IS RELATIONAL INTERVIEWING?

An interview is at the core of drama, fiction, real life. It is, in essence, the notion of a dialogue between two people.

Charlie Rose (Chautauqua Institution 2015)

For American journalist Charlie Rose, interviews lie at the heart of imagined and real life. For social scientists, they lie at the heart of much research. Through a variety of approaches, interviewing has produced insights into all manner of social and political phenomena. This book focuses on one specific approach—what I call “relational interviewing.” Relational interviewing is a method for generating data through interactions between researcher and interviewee. Its ethos is humanist. Its main ingredient is reflexivity. Its guiding principle is the ethical treatment of all participants. All three elements orient the researcher to interviewing as a learning process and to interviewees as people deserving of dignity and respect. As a learning process, interviewing enables the researcher to gain insight into participants’ worlds through interaction and dialogue. As people deserving of dignity and respect, interviewees are entitled to ethical treatment at all times, regardless of how likeable or unlikeable they turn out to be.

The pathway to learning and ethical awareness is reflexivity. By reflexivity, I mean a critical, ongoing examination of the way the researcher engages with others—be they participants, research assistants, interpreters, or other interlocutors. Reflexivity involves careful consideration of how issues of positionality—such as the researcher’s personal characteristics or theoretical vantage points—shape the research process. Such issues bear on the kinds of knowledge claims the researcher can advance. Reflexivity can also—and indeed should—involve developing an ethical sensibility that can attune the researcher to how her research design, practices, or strategies affect others. A reflexive disposition that includes a strong ethical sensibility will help to minimize harm to participants.

Methodological Underpinnings

Relational interviewing is based on an interpretivist rather than a positivist methodology.¹ The basic difference between these two modes of inquiry is the assumptions they make about the nature of the social world (ontology) and what is possible to know about that world (epistemology). Positivist methodology assumes that the social, like the physical, world is objectively knowable: the “truth” exists “out there,” waiting for researchers to “discover” it. An interpretivist methodology, by contrast, assumes that social phenomena do not claim any “real existence independent of how people think of them” (Schaffer 2016, Positivism and Interpretivism). The world is what people make of it. The meanings they give to “money,” “race,” or “witchcraft,” for example, constitute the very existence of these concepts. But for shared understandings about the worth of money, people would not work two jobs, play the stock market, or rob the corner store. Without historically situated understandings of what it means to be “black” and “white” in America, poor Irish immigrants would not have worked so hard at becoming white (Roediger 2007), and protesters today would not organize around the claim that “Black Lives Matter.” Without shared beliefs in hidden powers, early American settlers would not have punished those they deemed to be “witches” (Schiff 2015), and people across the world would not ascribe causality to witchcraft and sorcery (Ellis 2007; Ferme 2001; Schatzberg 2009).

An interpretive methodology assumes that explanations of these or any other empirical phenomena must start with an investigation into the meanings that people give to particular forms of social action and the social worlds and cultural forms these actions help to constitute. These meanings explain not only why people act the way they do, but also what it means, more generally, to “make” or “lose” money, to “be” white instead of black in America, or to “possess” special powers. The focus on meaning-making does not imply that there is no historical basis to these or any other social phenomena—to the contrary. It means that the occurrence of any event or historical moment, whether a war over beliefs or a revolution in technology, is not reducible to a single, objective, and unvarying truth. Instead, these events and moments are a matter of how people understand and make sense of them.

The armed conflict that Americans call “the Vietnam War,” for example, was, for some observers at the time, a justified effort to stop the spread of communism in Southeast Asia and, for others, a pointless exercise in killing and suffering. Today, many historians, journalists, and those who lived through the violence view the event very differently than how they saw it in 1968. Additionally, how Vietnamese then and now make sense of the war constitutes yet another set of historical truths about the same conflict. The very fact that in Vietnam, people refer to the event as “the American War” points not only to a very different way of understanding the conflict, but a very different starting point altogether. This war, like all others, is a matter of many truths, some complementary, some competing, and some whose emphasis has changed over time. It is these multiplicities of understanding that relational interviewing is well-suited to uncover.

Drawing on interpretivist assumptions, relational interviewing produces data that emerge dynamically through dialogue between researcher and interviewee. These data do not exist in free-standing form prior to the engagement between researcher and participant; rather, they are jointly produced through back and forth exchange. These interactions may be long or short, one-time or repeated, friendly or tense, casual or formal, or all of the above at different points in time as the research unfolds. No matter their duration or quality, they are always rooted in a specific social context, formed in part by “who” the interviewer and interviewee are, both individually and in interaction, the time of day, physical location, and presence or proximity of others. The larger context in which researcher and participant come together is also part of the interaction. Meeting right before key elections, during a severe drought, or just after financial collapse will also shape the kinds of interactions in which researcher and participant engage.

The data that interviewer and interviewee generate often take the form of narratives or stories that the latter tell about themselves and others. The value of these stories lies in the causal logics, worldviews, cosmologies, values, feelings, and shared understandings they reveal. Through the stories they tell, people locate themselves as agents in the various social worlds they identify with, aspire to, imagine, or inhabit. People’s stories provide insight into why they think certain events happened one way and not another, why perpetrators targeted certain families but not others (Malkki 1995), why some get ahead while others do not (Young 2004), and why a few risk their lives to rescue others, while most do nothing (Monroe 2004, 2011).

Practical Elements

In practice, relational interviewing starts with building working relationships, rather than rapport. Working relationships are negotiated between the interviewer and interviewee and are shaped by the interests, values, backgrounds, and beliefs that each brings to the exchange. These relationships enable the kinds of interaction that lie at the heart of the method. The elements that go into these exchanges are: active listening, learning to speak the language of interviewees, seeing “mistakes” as gifts, and treating participants with dignity and respect. I discuss each in turn.

Active Listening

Relational interviewing begins with active listening, which requires the researcher to take in multiple aspects of the interview at the same time, from smells and voices emanating from the next room to what the interviewee says and leaves unsaid through words, silences, and body language. It also entails noticing when interviewees tend toward embellishment, half-truths, or untruths, when they evade or avoid certain topics, and when they rely on rumors as sources of knowledge. All of these elements—half-truths, silences, rumors, and more—constitute “meta-data,” that is, data about data. Meta-data are important forms of information in their own right because words can hide just as silences can reveal (Fujii 2010). Stock answers or consensus accounts may reveal little, while fictional stories may say a great deal about the speaker’s aspirations and dreams (Portelli 1991).

Acquiring New Lexicons

In addition to active listening, relational interviewing requires the researcher to become familiar with the interviewee’s language or lexicon. Depending on the project, the researcher may need to study a foreign language, acquire new vocabularies, or hire an interpreter who can translate between the languages that she and the interviewee speak. Without gaining some familiarity with how participants talk about the world, the researcher will be at a decided disadvantage in trying to communicate with them. A humorous example of the problem with mismatched vocabularies comes from Bill Buford’s (1993) book about football hooligans in England. In this excerpt, Buford’s friend is trying to describe a Super Bowl game, the annual, professional championship match of the National Football League (NFL), which is based in the United States:

My friend—English, a writer, someone mindful of his language ...—was trying to describe the athleticism of Joe Montana [star NFL player in the 1980s and 1990s], but he didn’t have the vocabulary. He didn’t know what a line of scrimmage was; pass, bomb, wide receiver, third and long: the terms, so firmly established in the linguistic reserves of anyone who has grown up with “American” football, were from a foreign language that my friend hadn’t yet mastered. He grew desperate and dipped deep into the archives of his own sports-page vocabulary and came up with long ball, pitch, fast bowling, crease—a chaotic combination of terms from English football, cricket and (for all I know) croquet as well.

(Buford 1993, 315–16; emphasis in original)

In relational interviewing, the researcher might start out like Buford’s friend, the Englishman who lacks the vocabulary to talk American football. Rather than impose English terms onto a non-English sport, however, the researcher would be better served learning how to “speak” professional football as it is played in the United States.

Having no familiarity with the interviewee’s language can lead to misunderstandings and misinterpretations of what people are saying, as Andrew Herod (1999, 318) discovered while researching labor unions in Eastern Europe. Concepts whose meanings he took for granted while interviewing in the US, such as “collective bargaining” or “labor relations,” carried different meanings in the Czech² and Slovak Republics because of those countries’ own histories and what was then a recent transition from communism to market-based economies. As Herod (1999, 318) concluded, the “language” he spoke with North American trade union officials “did not seem to make much sense in the context of Eastern Europe.”

Without some familiarity with the “local language,” the researcher also risks being unintelligible to participants. Carol Cohn (1987), for example, studied consultants on nuclear weapons systems who worked for the US Department of Defense. These “defense intellectuals,” as she calls them, spoke a very specialized language that gave no hint of the magnitude of destruction the weapons were capable of unleashing. At first, Cohn tried to speak to these men in plain English, even though she had “acquired proficiency in the[ir] techno-strategic jargon.” By not speaking in their language, notes Cohn (1987, 708), the men refused to talk back; instead, they treated her as if she were “ignorant, simpleminded, or both.”

Having no familiarity with the interviewee’s language might also lead the researcher to offend potential participants, as Belinda Robnett found when she began interviews for her dissertation. To prepare for her first interview, Robnett drew up a list of questions using the analytic terms with which she was most familiar. One was “sexism.” When Robnett telephoned the interviewee to ask about sexism in the US Civil Rights movement, the woman so strongly objected to use of the term that she declined to continue with the interview. But before hanging up, the interviewee, a black woman, explained that sexism was not a word used during the Civil Rights era and that, furthermore, the term more readily applied to white, not black, women’s experiences. Far from being the victim of “sexism,” the interviewee felt empowered by her participation in the movement. This first interview proved invaluable to Robnett’s dissertation project, which eventually became a book entitled How Long? How Long? African-American Women in the Struggle for Civil Rights (Robnett 1997). Through this first attempt at interviewing, Robnett gained a more nuanced understanding of black women’s experiences in the Civil Rights movement and became sensitized to the importance of historical context.³

Learning through Missteps

Viewed from the perspective of relational interviewing, Robnett’s initial attempt at interviewing and Cohn’s efforts at talking to defense intellectuals in plain English do not constitute mistakes. Both exchanges were valuable because they imparted knowledge the two researchers may not have gleaned any other way. Through a single interaction, each scholar learned the importance of not imposing her own vocabulary onto the experience of others, but rather allowing interviewees to use their own language. Both also learned that language—whether in the form of specific words or specialized vocabularies—can have important meaning to interviewees that goes beyond the functional requirements of communicating with another person. Robnett’s interviewee, for example, may have objected to the term “sexism” not only because it referenced a class of women who never experienced racial injustice, but also because it dismissed the very possibility that black women exercised both power and agency during the Civil Rights movement. For Cohn’s interviewees, some may have insisted on speaking in the techno-jargon of their profession—even to a civilian researcher—because of the constitutive power of specialized “talk.” Fluency in the language of nuclear weaponry may have been an important signifier of belonging; not speaking the language would have meant relinquishing their place in a rarefied world.

This brief analysis underscores that “mistakes” do not just teach researchers what not to do; they can also provide insights that may not have emerged had the “right” words or language been used from the beginning. By reflecting on what these “failed” interactions reveal, researchers can turn moments of regret into gifts of valuable insight.

Treating People with Dignity and Respect

The overarching principle that guides relational interviewers in their interactions with participants is to treat everyone as “ends” in themselves and not as a “means” to some other end, such as a book or dissertation. In more concrete terms, it means treating all interviewees with the same dignity and respect regardless of how forthcoming or evasive they may turn out to be. Treating people as ends, not means, requires developing a heightened ethical sensibility that can alert the researc...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Dedication
• Table of Contents
• Series Editors’ Foreword
• Preface
• Acknowledgments
• 1 What is Relational Interviewing?
• 2 Building Working Relationships
• 3 Selecting, Finding, and Approaching Interviewees
• 4 Strategies for Conducting Interviews
• 5 I Have My Data—Now What?
• 6 The Ethos of Relational Interviewing
• Appendix: Interview Excerpts
• References
• Index